MoodLogic had assembled a rich database of music metadata that contained quantitative (artist, album, genre, etc.) as well as qualitative (mood, song similarity, etc.) information about millions of songs. One of the ways the company demonstrated the potential uses of this data was via streaming internet radio applications that managed to be personalized while still obeying DMCA rules for internet radio playlists.

Some of our radio apps enabled the user to create a channel by selecting a set of attributes such as genre, tempo, or mood, while others would build custom channels around a favorite song or artist. If that sounds a lot like the now-popular Pandora, that’s because it was… except that Pandora came along later. Our radio apps also used AJAX-y techniques to asynchronously communicate with the server well before that term existed.

My responsibilities included:

• Feature refinement
• Interaction design
• Front-end development (HTML, javascript, and Flash)
• Managing other developers

It was my pleasure to collaborate with Dave Vanderkloot of 1530 Design on the Flash version of our radio app, and with Tobias Scheck of Topazmedia on the DHTML version.

MoodLogic had assembled a rich database of music metadata that contained quantitative (artist, album, genre, etc.) as well as qualitative (mood, song similarity, etc.) information about millions of songs. After other means of commercializing the company’s technology in the post-crash era failed, we embarked on a last-ditch effort to generate revenue — by building a jukebox application that leveraged our unique technology and offering it for sale to end users.

Our plan seemed like a longshot, given the ubiquity of free jukebox apps (iTunes, WinAmp, Windows Media Player, etc.), and yet it worked. MoodLogic won accolades and paying customers for its unique features packaged in a slick and usable UI.

Here are a few things reviewers have said about the app:

“It is an amazing product… The options are limitless and have saved me hours of sifting through my collection to make playlists.” - Music Gear Review

“Versatile tool for music organization and management. Very smooth interface.” - Winamp.com

“Attractive interface… We love the MoodLogic concept.” - ZDNet Reviews

My responsibilities on this project included:

• Feature refinement
• Interaction design
• Functional specifications and wireframes
• Usability analysis
• Visual design

I left MoodLogic in 2001, and the company continued to refine the product through late 2003, though the basic design never changed. The company’s founders were anxious to sell out, and sadly they seem to have abandoned their loyal customers by early 2004. Though you wouldn’t know it from their website, MoodLogic was acquired in 2006 by competitor AllMediaGuide.

I began my work at Apple translating designs into HTML emails to be broadcast en-masse to Apple customers through an internally-developed system named Chatterbox. As happens with many internally-developed applications, over the years Chatterbox had far outgrown its original scope in terms of features, number of worldwide users, and the sheer volume of emails being broadcast to customers. This had all happened with no thorough analysis of the usability of its user interface — Apple’s business had been exploding quarter after quarter, and most efforts had been put into making the system cope with an ever-increasing throughput of emails.

Using Chatterbox day after day, the interface became increasingly frustrating to me and also to my fellow HTML coders in Cupertino. The UI was simply a mess, and it was slowing us (and dozens of Apple employees around the world) down and contributing to delays and mistakes in critical corporate communications (you can’t un-send an email, so any mistakes in outgoing messages are embarrassing… especially when they’re seen by millions).

Management agreed that the UI needed to be redesigned, and decided that the imminent 2.0 release, when many significant changes were being made under the hood and a new security model also had to be implemented, was the best time to do it. I proposed a sweeping reorganization of the UI and, to my surprise, Jason Winchell, the main backend UI developer on the project, committed to making it happen.

Over the next several weeks, Jason and I worked to completely rebuild the system’s UI — to make it immediately familiar to anyone who’s used iTunes or Mail, to reduce the number of clicks and page refreshes it takes to get things done, to enable searching and sorting through thousands of email broadcasts, to make it harder to make mistakes, and to accommodate the new, business-mandated security model with its myriad of user roles and permissions. The end result is a system that’s much easier on the eyes, faster to use, and will scale as new features are added.

My responsibilities on this project included:

• Interaction design
• Usability analysis
• Wireframes and mockups
• Visual design and graphics production
• XHTML/CSS/Javascript coding
• Training materials and system documentation

This was a challenging but rewarding project, as we were able to significantly speed, simplify, and de-frustrate-ize the daily work of dozens of hardworking Apple employees around the world — work which is very important to Apple’s bottom line.

This UI redesign would not have been successful without Jason’s amazing work, but I also have to thank management (Greg and Chris) for their support, Grant and Jaya for their dedicated testing, and especially Ben Allen and Joe To for contributing valuable design & feature ideas, helping a lot with the XHTML/CSS development & debugging, and for graciously taking over my day-to-day responsibilities so I could focus on this project.